Language, cognition, interaction : Conceptual blending as discursive practice

Abstract

Conceptual Blending and Mental Spaces Theory (MSCI) has been proven to be a highly appealing theory for describing the cognitive processes which give rise to creative language usage. By bringing distinct semantic scenarios into mutual interaction, novel structure emerges which is fundamentally more than the sum of its parts and contains elements that are found in neither of the input spaces. Despite its intuitive attractiveness and promising ideas, MSCI has been criticised for a number of reasons. Most prominently, it has been accused of losing sight of the phenomenon, neglecting the historicity, material and social dimension of situated cognising, and therefore not being able to account for contextualised meaning construction. MSCI thus appears to be largely rooted in ideational theories of meaning which neglect the social and action-based dimension of meaning-making (as emphasised for example by Wittgenstein (1953)). MSCI is furthermore considered to be simply too complex, proposing artificial, post-hoc constructs on seemingly isolated minds. A case in point is the highly complex blending typology that lies at the heart of MSCI. This typology proposes a continuum of frame-integration complexity which cognisers have at their disposal; this ranges from simplex blends to double-scope blends. The authors claim that double-scope frame integration is the very factor which makes human achievements such as mathematics, arts and religion possible. Yet the typology proposed has so far not been tested on actual data, and has rightly been accused of constituting a post-hoc analytic category. MSCI s focus thus seems to lie on competence rather than performance despite aspiring to defy the semantic-pragmatic distinction (cf., e.g. Fauconnier, 2006). A usage-based approach to language and Cognitive Linguistics in general claims to be rooted in the tradition (cf. Geeraerts, 2005) should, however, be able to overcome this rigid Chomskyan dichotomy (Mukherjee, 2004:96) between theoretical ability and genuine behaviour, and propose a model of language cognition [ ] able to account for actual usage (ibid:87). For this reason, this study aims to investigate blending-in-interaction in order to a) trace the way in which conceptual blending is negotiated by interactants in situ, and b) determine the extent to which participants-in-interaction orient to the allegedly post-hoc blending typology proposed by MSCI s founding fathers.In order to avoid the ubiquity problem and the often-problematic delineation of Mental Spaces, one phenomenon that involves two semiotically distinct semantic scenarios is studied in detail. Impersonation humour requires the appropriation of voices other than one s own, and may therefore be considered to be an excellent example of conceptual blending that occurs on various levels of frame integration complexity from mirror- to double-scope blends. Blending-in-interaction is studied in the methodological framework suggested by Ethnomethodological Conversation Analysis (EMCA). The results of the analysis show that MSCI cannot be considered a theory of meaning-making that transcends the time-honoured division into semantics and pragmatics. Despite claiming that MSCI is able to account for how meaning is constructed and how language prompts for meaning (Fauconnier & Turner, 2002:147), the meanings resulting from the integration of distinct semantic scenarios do not relate to the process of conceptual integration. The meanings emerging from the joint coordination of frame integration are primarily guided by jointly negotiated assumptions regarding the language game currently being negotiated, and by the relevance which a given sign is taken to hold in relation to that particular language game. For MSCI to be a theory that transcends the semantics-pragmatics division and that is able to account for fully contextualised meanings, it will need to incorporate notions of Common Ground, language games, intention and relevance. It will also need to explicitly acknowledge the historicity inherent in meaning-making processes. I propose a model that takes a serious stance as to the key role that beliefs play regarding contextual relevance in online, interactional meaning-construction. By incorporating notions of signs, stimuli, relevance, language games and Common Ground and regarding conceptual blending as a cognitive tool, it hopes to be able to account for the emergence of fully-contextualised meanings arising from jointly negotiated frame integration. The model thus accounts for the pivotal role which the pragmatic plays in meaning-making. Thus, that which been deemed irrelevant (cf. competence performance distinction) in previous research in cognitive sciences is actually revealed to exert considerable influence on cognition.